Editing databases with Vim/Perl/DBI
Perl's Data-Base-Independent (DBI) module provides programming language level access to a lot of databases. Vim hosts an embedded Perl interpreter. So it is only a matter of some key strokes to interactively issue DB commands from within Vim or to search, edit, and replace database contents including retrieval and storage. Of course "create table" scripts can be worked upon in Vim as well as storing recurring patterns in Vim functions or Perl modules. Prerequisites: Vim needs to be compiled with Perl support enabled. See the |if_perl.txt| manual page! The CPAN module DBI as well as an appropriate database driver has to be installed with Perl in order to execute these Vim commands: " connect to perl's dbi module: :perl use dbi; " connect to the database: :perl $dbh = dbi->connect( "DBI:mysql:$DBNAME:$HOST",$USER,$PASSWORD, { raiseerror => 1}); " perform a simple query: :perl $result = $dbh->selectall_arrayref("show tables;"); " insert the list of tables into the current buffer's top: :perl $curbuf->Append(0, map($_->0, @{$result})); In MySql the command "show tables;" results in a list of table names. Inserted into a Vim buffer this results in one line per table. You can find more on my web page http://members.chello.at/intelliware/dbEdit Comments When I have a some SQL in vim that I want to see the results of, I use this simple technique to send it to Postgres: 1. Highlight the the SQL. 2. :!psql -d my _database (for Postgres) Vim will filter my lines through that command replace them with the result set. When I'm done reviewing the results, I use "u" to undo the change. If I repeat this cycle, I can use use :CTRL-P to get back the previous command to use, and instead of typing it again. For larger pieces of SQL, I put it in a file and use: :!psql -d my_db -f % to send the whole script to Postgres. ---- Running an external process is the way Emacs handles database queries as well. Cursors and other database programming are not possible this way. Perl's DBI module provides the Vim user with full interactive access to the database in the same way as used by graphical user interfaces and application programs. Also the single process solution (editor + database access in one process) opens only a single permanent database connection which is much quicker and allows bigger transactions. Similar motivations led to the programming language Tcl: Why should one embed a command interpreter into application programs? The answer after many years of Tcl is given by a lot of scripts weaving together all kinds of programs as well as Tcl's graphical user interface library Tk which is also used for Perl. For example Linux consists on one hand of many (text) files, which can be edited by Vim, and on the other hand of many complex relations between them, which can be handled effectively by a relational database. Merging Vim's many facilities with those of databases makes it possible to build a simple yet effective digital library with only the means at hand. ----